[code-review] Tie::Hash::Expire

(This is a second transmission of this message with a different 
subject.  Another poster suggested to me that perhaps it isn't the best 
idea to mimic a common virus / worm.  ;-)  I'm sorry for those of you who 
will see it twice.)

Hi all,

	I just became aware of this list a few minutes ago.  What a great idea!

	Yesterday I uploaded the first version of my first CPAN module, 
Tie::Hash::Expire, a package providing hash keys that expire and cease to 
exist after a user-set period.  I would be much obliged if anyone here 
would take a look.
	As I'm new to distribution through CPAN, I'd particularly appreciate any 
comments about how I might have messed up there.  If the distribution looks 
non-standard in some way, doesn't work with your CPAN shell, or is in some 
other way funny, it is probably my oversight / error rather than 
intentional, so I would prefer to hear about anything like that.
	Also, if anyone is motivated to really dig into this a little, I'd very 
much appreciate comments on the internal data architecture.  When I 
originally wrote this, I implemented the data as a normal hash with values 
of an array reference containing a timestamp and the value for each 
key.  This is enormously simpler than what I ended up doing.  I made the 
switch because I felt that being able to eliminate expired keys through a 
binary search might represent substantial savings.  I'm not any longer sure 
that I I'm gaining more than I'm spending.  One advantage to this approach 
is that FIRST/NEXTKEY return the keys in order of insertion / 
modification.  I haven't documented this feature as I'm still a little in 
the air about whether I should retain the implementation that 
coincidentally makes that possible.

Cheers,
Jeff